We pioneered the concept of signal transduction therapy with CAI and have completed our phase II study of single agent CAI in ovarian cancer and our combination study with paclitaxel. One patient remained on CAI/paclitaxel for over 4 yrs before progression of disease; a manuscript is in preparation. We have expanded our approach to signal transduction therapy and have applied microproteomic techniques to the analysis of signal pathway modulation in tumor samples. Microdissection and proteomic tissue lysate array profiling are ongoing to assess the clinical effects of two molecularly targeted agents in the tumor and stroma of women with recurrent ovarian cancer. We have optimized techniques for acquisition of sentinel tumor lesions and have obtained biopsies prior to initiation and at 1 mo of therapy. Stroma and tumor are microdissected and subjected to protein array to analyze the initial and downstream activity of regulation of key nodes in these pathways. Preliminary tissue lysate array results from the imatinib study shows that PDGFR and c-kit are expressed in more abundance and more frequently activated than expected, and c-kit is modulated in tumor and stroma, with no correlation to clinical outcome. Accumulation of ascites and pleural effusions on our imatinib protocol was unexpected and severe in a subset of patients; we found a strong association with increased concentrations of plasma VEGF, PDGF-AB, PDGF-BB, and IL-6 in those patients, strongly suggesting the imatinib therapy upregulated cytokine production. Preliminary results from the gefitinib study indicate that changes are measurable in total and activated AKT and ERK, without clear correlation with tumor drug sensitivity; results are ongoing to measure changes in phosphorylation of 4 EGFR phosphotyrosine endpoints. Initiation of accrual of a phase I/II trial of sorafenib and bevacizumab, combinatorial anti-angiogenic anti-signaling therapy has begun. Preliminary results indicate activity in ovarian cancer patients. Cohort 2 of this trial incorporates serial biopsies, DCE-MRI, and PET scans to assess biological and biochemical activity of this combination. A phase II trial of this combination for ovarian cancer patients is under development. The tissue lysate array approach creates the opportunity to evaluate in situ the effects of circulating drug on the putative tumor and stromal targets to test microenvironment interaction and molecular target hypotheses. Ovarian cancer serum proteomics is advancing with a multi-institutional trial for the collection of serum samples from women in first remission of advanced stage ovarian cancer. In addition, collection of serum from which to develop a proteomic signature of malignancy v. benign disease from women undergoing surgical assessment of a pelvic mass is about to initiate accrual through the Gynecologic Oncology Group (Kohn, PI). Serum samples from these two studies will be used for training and blinded validation of serum proteomic signatures of disease recurrence and malignancy, respectively. Thus, we have translated both exciting proteomic technology advances to direct clinical investigation.